Parallel for AI agents

Parallel's Task API turns natural language research questions into structured, schema-compliant web research results. Analysts and product teams use it to automate multi-step web research. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.

Connect Parallel See the 41 actions ↓

Web data and scrapingAPI keyMCP + RESTparallel.ai

AUDIT LOG · PARALLELPOLICY: acme-support

09:41:07 · claude · u_8f2read

parallel.list_monitors

List Monitors✓ allowed · 212ms

09:41:08 · claude · u_8f2read

parallel.search

Parallel Search✓ allowed · 164ms

09:41:09 · claude · u_8f2write

parallel.create_monitor

Create Monitor✓ approved · approved by user

09:41:12 · claude · u_8f2destructive

parallel.delete_monitor

Delete Monitor✕ blocked · policy: deny

EVERY PARALLEL CALL, ON THE RECORD

01 · USE CASES

What agents do in Parallel.

Run a structured research task

Submit a research question and get back a schema-compliant answer sourced from the web.

Check a monitor's recent events

List monitor events read-only to see what changes a tracked query has picked up.

Extract content from a set of URLs

Pull clean content from a batch of URLs to feed into a research summary.

02 · ACTIONS

41 Parallel actions, graded by risk.

Every Parallel action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.

read

Look things up. Allowed by default.

parallel.find_all
Start FindAll Run
parallel.search_web
Search Web
parallel.get_monitor
Get Monitor
parallel.list_monitors
List Monitors
parallel.get_task_group
Get Task Group
parallel.list_monitor_events
List Monitor Events
parallel.list_task_group_runs
List Task Group Runs
parallel.list_task_run_events
List Task Run Events
parallel.list_current_monitors
List Current Monitors
parallel.get_find_all_run_result
Get FindAll Run Result
parallel.get_find_all_run_schema
Get FindAll Run Schema
parallel.search
Parallel Search
parallel.extract
Extract Content from URLs
parallel.retrieve_memory
Retrieve Memory
parallel.retrieve_monitor
Retrieve Monitor
parallel.retrieve_task_run
Retrieve Task Run

+ 8 MORE

write

Create and change things. Allow, or ask the user first.

parallel.create_monitor
Create Monitor
parallel.update_monitor
Update Monitor
parallel.create_task_run
Create Task Run
parallel.create_task_group
Create Task Group
parallel.add_runs_to_task_group
Add Runs to Task Group
parallel.create_chat_completions
Create Chat Completions
parallel.add_enrichment_to_find_all_run
Add Enrichment to FindAll Run
parallel.suggest_task
Suggest Task
parallel.simulate_event
Simulate Event
parallel.extend_find_all_run
Extend FindAll Run
parallel.ingest_find_all_run
Ingest FindAll Run
parallel.stream_find_all_events
Stream FindAll Events
parallel.stream_task_run_events
Stream Task Run Events
parallel.stream_task_group_events
Stream Task Group Events
parallel.generate_grounded_response
Generate Grounded Response

destructive

Delete, cancel or archive. Ask first, or deny outright.

parallel.delete_monitor
Delete Monitor
parallel.cancel_find_all_run
Cancel FindAll Run

03 · HOW IT WORKS

Parallel in three steps.

01Your users connect ParallelThey add their Parallel api key on Arc0 Connect, under your brand. It goes straight into the vault.
02You set the rulesReads run, writes like “create Monitor” can wait for the user, and “delete Monitor” can be denied outright.
03Any agent can actYour agent calls Parallel through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.

POLICY.TS

await arc0.policies.set('parallel', {
  read: 'allow',
  write: 'ask',        // create_monitor
  destructive: 'deny',  // delete_monitor
})

# Claude Code: the same connection, one URL
$ claude mcp add --transport http arc0 \
    https://mcp.arc0.ai/u/u_8f2

04 · AUTH AND DATA

How Parallel connects.

Users add their Parallel api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.

The same Parallel connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →

AUTH: API key
CREDENTIALS: Per-tenant encrypted vault
MODEL SEES: Results only, never credentials
AUDIT LOG: Every call, on every plan

05 · WORKS WITH

Use Parallel from any agent.

Claude

ChatGPT

Cursor

Codex

VS Code

OpenAI Agents SDK

Claude Agent SDK

Vercel AI SDK

Mastra

LangGraph

06 · RELATED

More web data and scraping apps.

Just One API288 actions

Apify183 actions

ALL WEB DATA AND SCRAPING APPS →

07 · FAQ

Parallel and Arc0, answered.

Q01

Can I use Parallel with Claude, ChatGPT or Cursor?

Yes. Connect Parallel to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the Parallel actions you allow.

Q02

How do users connect Parallel?

Users add their Parallel api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.

Q03

Which Parallel actions can my agent take?

41 in total: 24 read, 15 write and 2 destructive, such as “create Monitor”. Your policies decide which of them each agent may call.

Q04

Can I stop my agent from deleting things in Parallel?

Yes. Actions like “delete Monitor” are graded destructive. Set destructive actions to deny, or to ask so the user approves each one, and blocked calls still show up on the audit log.

Q05

Can my own backend call Parallel too?

Yes. The same Parallel connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.

Get started

Plug Parallel into your agent.

Your users connect Parallel once, under your brand. Your agent gets 41 actions behind your policies, with every call on the record.

Connect Parallel Browse all apps →

Free to build · MCP + REST · Audit log on every plan